[Openswan Users] Tunnel breaks after longer connection problems

[MarekGreško]

Hallo,

I use openswan 2.1.2 with recent fc2 2.6 kernel and I run into problems 
bringing up tunnel after longer Internet connection problem. Our Internet 
connection was down for 2 days. Gateway 1 was rebooted during these 2 days. 
After bringigng Internet connection up the ipsec had problem reestablishing 
tunnel. On the gateway 2 I found these logs repeating very frequently:

Jul 18 04:02:03 gw pluto[1631]: | *time to handle event
Jul 18 04:02:03 gw pluto[1631]: | event after this is EVENT_RETRANSMIT in -16 
se
conds
Jul 18 04:02:03 gw pluto[1631]: | handling event EVENT_RETRANSMIT for 
193.87.3.8
2 "michal-pio" #732575
Jul 18 04:02:04 gw pluto[1631]: | inserting event EVENT_RETRANSMIT, timeout in 
4
0 seconds for #732575
Jul 18 04:02:04 gw pluto[1631]: | next event EVENT_RETRANSMIT in -17 seconds 
for
 #732512
Jul 18 04:02:04 gw pluto[1631]: |  
Jul 18 04:02:04 gw pluto[1631]: | *time to handle event

After a while....
Jul 18 04:02:08 gw pluto[1631]: | peer and cookies match on #732406, provided 
ms
gid 64e56dde vs 1f0c984f
Jul 18 04:02:08 gw pluto[1631]: | peer and cookies match on #732405, provided 
ms
gid 64e56dde vs 193427b3
Jul 18 04:02:08 gw pluto[1631]: | peer and cookies match on #732404, provided 
ms
gid 64e56dde vs 15a560a1
Jul 18 04:02:08 gw pluto[1631]: | peer and cookies match on #732403, provided 
ms
gid 64e56dde vs 5ce97217

.... and .....

Jul 18 04:02:46 gw pluto[1631]: | peer and cookies match on #730468, provided 
msgid 64e56dde vs 64e56dde
Jul 18 04:02:46 gw pluto[1631]: | state object #730468 found, in 
STATE_QUICK_I1
Jul 18 04:02:47 gw pluto[1631]: | our client is subnet 192.168.10.0/24
Jul 18 04:02:47 gw pluto[1631]: | our client protocol/port is 0/0
Jul 18 04:02:47 gw pluto[1631]: | peer client is subnet 193.87.32.0/24
Jul 18 04:02:47 gw pluto[1631]: | peer client protocol/port is 0/0
Jul 18 04:02:47 gw pluto[1631]: | install_ipsec_sa() for #730468: inbound and 
outbound
Jul 18 04:02:47 gw pluto[1631]: | route owner of "michal-pio" erouted: self; 
eroute owner: self
Jul 18 04:02:47 gw pluto[1631]: | could_route called for michal-pio 
(kind=CK_PERMANENT)
Jul 18 04:02:47 gw pluto[1631]: "michal-pio" #730468: ERROR: netlink response 
for Add SA esp.17ba61ff at 195.49.191.34 included errno 3: No such process
Jul 18 04:02:47 gw pluto[1631]: | state transition function for STATE_QUICK_I1 
had internal error
Jul 18 04:02:47 gw pluto[1631]: | next event EVENT_RETRANSMIT in -39 seconds 
for #730561
Jul 18 04:02:47 gw pluto[1631]: |  
Jul 18 04:02:47 gw pluto[1631]: | *time to handle event

I got around 300 MB of logs on both gateways in 2 days.

What could be the problem? Only rebooting of both gateways helped.

Any help appreciated.

Marek